By lowering cost and not investing profit to the company? Yes, short-term v long-term, but who in this world cares about anything after their next salary?
Lovely, so yet another promise to federate which will never materialise! Still going with the Drew’s reply in https://is.gd/5wwQy2 (yes, two years old, and he slightly softened his stand since then):
> SourceHut is already federated via email. We have no intention of adding ActivityPub support at this time.
Federated repositories is something very similar to paperless office, distributed authentication (OpenID), and distributed computing … it has been promised since forever, and nobody has ever seen it in the real life, and even less supported by somebody who matters. And yes, those who matter don’t help by sabotaging any efforts towards it.
You said SHOULD. Yes, I absolutely agree that politicians (and I very intentionally do not call any names) should be criminally punished most harshly for abusing their position for personal enrichment or for some other egoistical goals. On the other hand, these are the people we, as totality of all voters, chose for their function. The main punishment for a politician in democracy should be the threat of losing next elections, not criminal prosecution. And of course, per definition, in every democracy every politician has a majority of citizens, who considers him stupid and in the hysterical environment of the current political life (hysterical for many more or less good reasons) such politician is not only opponent, but an enemy if not a traitor. There is an unfortunate tendency to convert this adversarial feeling into full blown hate and accusations of criminal misconduct.
I didn’t mean to be pedantic. My experience with Jujutsu was as bad as with Mercurial (and hg-git, I believe?) … after couple of hours working with my repo (https://git.sr.ht/~mcepl/m2crypto), looking at it with git, I got incomprehensible mess of destroyed multiple unmergeable heads (especially for Mercurial), and completely destroyed other branches, where I was not planning to work (Jujutsu). The only resolution in both case was export my work to plain patches, rm -r the checkout, clone again with git and forget anything about those other VCSes.
So, I claim that Jujutsu actually doesn’t work well with git repositories (and forges) very well, and I would like to see a native one.
yes, I haven’t managed for a week, but I tried to Jujutsu and gave up. It is too complicated for no visible gain. Plus, there is no infrastructure supporting it (in the end, you store the stuff in the crippled git repositories). `git commit --amend` and `git rebase --update-refs` together with few scripts (e.g., https://git.sr.ht/~mcepl/git-fixup) does the same, and I am still with true git.
I don’t think it is that simple. Itanium was for years supported for example by RHEL (including GCC working of course, if anybody cared enough they could invest into optimising that), it is not like the whole fiasco happened in one moment. No, Itanium was genuinely a bad design, which never got fixed, because it apparently couldn’t be.